The Heisenberg Principle of Public Policy
There’s a tendency among wonks, pundits, journalists, academics, and members of the political minority to favor more theoretically “pure” policies than are advanced by those in power. Those who have the actual responsibility of governing do not often try to achieve doctrinal purity, because they do not have the luxury of ignoring public attitudes (skepticism, mistrust, fear, etc.) about the effects of implementing the policy.
Many critics of the Democrats’ health reform plan think we need to radically shake things up—drastically cut spending; restructure incentives for both consumers and providers; end the employer-based insurance system; institute single-payer over night; etc. Matt Yglesias chastens proponents of the Wyden-Bennett approach (which would end the tax exclusion for employer-provided health benefits) for excesses in this vein:
Wyden-Bennett fans in the commentariat…have developed a habit of massively understating the transformative potential of the Obama administration’s health care proposals. It’s true that Obama shies away from a Wydenesque sudden shock to the employer-based system. But Obama does that because even though journalists and think tankers tend to favor such a sudden shock the public hates that idea and is terrified of it and nothing any politician says is going to change that. What Obama’s plan does accomplish, however, is to set the stage for a phased transition to a system in which employers are cut out of the loop.
Shock-and-awe punditry is entertaining and useful for framing issues in public discourse, but the responsibility of actually governing changes how policy is evaluated, as feasibility conditions and dynamic public attitudes feed back into the substantive development of the policy. This amounts to a kind of political Heisenberg effect, whereby public perception of a policy may alter its substance and its merit as policy.
The upshot is that, even when radical change is called for, phased transitions are really the only prudent course.1
- Probably. [↩]

